AIFORYA Sitemap

1.3.3

• The sitemap no longer lists archives that cannot be reached. A post type or a taxonomy was
  included as long as it was registered, which is not the same as being viewable. WooCommerce,
  for instance, always registers its product attribute taxonomies (pa_...); unticking “Enable
  archives” only stops their archive pages from resolving. A sitemap saved before that change kept
  submitting those addresses, and every one of them returned a 404.
• Both post types and taxonomies are now filtered with is_post_type_viewable() and
  is_taxonomy_viewable() — the same functions the WordPress core sitemap uses.
• Nothing to do on your side: your saved settings are untouched, and any archive that is reachable
  stays in the sitemap exactly as before.

1.3.2

• WooCommerce product attributes (pa_...) are no longer added to the sitemap by default. These are faceted-navigation archives — “diameter = 12”, “colour = red” — not pages a shop wants ranking in search results. Measured on a live store: 339 of the 449 submitted addresses (75%) were attribute archives, and 24 of them contained only one or two URLs. Crawl budget was being spent on near-empty pages instead of the product pages.
• Product categories, brands and tags stay included, exactly as they have since 1.2.0 — it was their absence that was fixed then, and nothing about that changes.
• A shop that wants its attributes in the sitemap keeps them with one line: add_filter( 'aiforya_smap_inclure_attributs_produit', '__return_true' );
• Existing sites see no change: saved settings are kept untouched. The new behaviour applies to fresh installs only, and the taxonomies remain editable in the settings screen.

1.3.1

• Fix: the home page was listed twice in the sitemap on any site using a static front page. Search engines were being handed the same address as two separate entries, which wastes crawl budget and can dilute how the page is understood.
• The home page is now listed once, and it carries its real last-modified date instead of none.

1.3.0

• The sitemap is now declared in robots.txt. Until now the plugin generated the file, the file answered correctly… and no search engine knew it existed: you had to submit the address by hand in Search Console. This is now automatic.
• The line is only added when the sitemap actually has something to announce: it is omitted if the plugin is switched off in its settings, if the site asks not to be indexed, or if nothing is published yet. Announcing an empty address is worse than announcing nothing.
• WordPress’s own Sitemap: line is neither removed nor altered — both coexist, as the robots.txt standard allows.
• A site with nothing published now returns 404 on /aiforya-sitemap.xml instead of an empty index served with 200. An empty index is a valid document that a search engine reads as “this site has nothing left”, and that de-indexes.

1.2.0

• The sitemap is now a paginated index, as the sitemaps.org standard prescribes. /aiforya-sitemap.xml lists /aiforya-sitemap-posts-<type>-<n>.xml and /aiforya-sitemap-taxonomies-<taxonomy>-<n>.xml files of 1,000 addresses each.
• Important fix: past 2,000 addresses the sitemap was truncated — and the cut was made from the end, so a shop with 2,000 products lost all of its categories, with no message at all. The number of addresses is no longer capped.
• A page requested out of range now returns 404 instead of an empty sitemap.
• Addresses are sorted by ascending id: an address stays on the same page as long as none is added or removed, instead of moving on every content change.
• Rewrite rules are refreshed automatically on update.

1.1.0

• All public taxonomies can be included, not only categories and tags. On a WooCommerce shop, product categories, brands and attributes were missing from the sitemap with nothing to signal it.
• New screen to choose which taxonomies to include.